WIXLIB

DNA testing and analysis Final DNA testing and statistical analysis was then conducted at the Thünen Institute inGermany to determine the effectiveness of the approach to assign samples taken fromwarehouses and processing facilities back to PAUs and harvest zones.2.3 Risks and assumptionsThe key risk identified during the proposal phase was security in East DRC. At the time theGovernment of DRC had recently negotiated a peace deal with rebels in the north and SouthKivu, allowing for field work to take place. Sampling in this area was in fact disrupted, with abatch of bark samples confiscated by local militia, and repeat sampling undertaken.A further risk was possible problems of DNA extraction from Prunus bark which would requireadditional efforts to develop genetic markers. This risk was mitigated through a trial DNAextraction activity at the beginning of the project, and successful analysis against existinggenetic markers of Prunus persica.Assumptions were made that there would be no problems associated with the transportationof bark samples from Cameroon and DRC to the laboratory in Germany, or with existingharvest and processing procedures that would allow for document matching of bark back toindividual trees.6

3PROJECT PERFORMANCEBased on the performance levels detailed below and presentation of project deliverables,DoubleHelix considers the project objective of building the capability to trace Prunus africanabark back to specific PAUs or harvest zones to a high level of confidence to be achieved.With the adjustment to the scope of work, the original objective has been exceeded,considering it is now possible to assign samples back to both individual trees and populations.Either approach can now be applied, depending on which approach is practical in eachcountry or supply chain.3.1 Achievement of project outputs and activitiesThere was some variation between the original planned outputs and activities which resultedin the successful completion of the project. A comparison between the project outputs andactivities as originally devised and the final outcomes follows:Output 1.1: Development of genetic markers for Prunus africana suitable for DNAfingerprinting of bark (differentiation of bark between individual trees of the samespecies).Activity 1.1.1 Trial DNA extraction from bark samplesProjectdocumentdescriptionInitial tests will be performed on Prunus africana bark samples toconfirm that DNA of sufficient quality and quantity can be extracted frombark on a routine basis.Changes toactivityRealized activityNoneComplete. A preliminary test to verify that DNA can be extracted from P.africana bark was successfully completed. Five samples were submittedto the laboratory for testing and DNA of sufficient quantity and qualitywas extracted from all five. The results of this DNA extraction test ispresented in Appendix A.Activity 1.1.2 Population samplingProjectdocumentdescriptionTwo hundred (200) samples will be collected from a number of differentpopulations within the species range. Twenty (20) samples each from 5different populations across Cameroon and another 5 differentpopulations in DRC will be collected.Changes toactivityThis activity was extended with additional collections in DRC to buildcapability for assignment to population as well as individual matching.Realized activityComplete. A total of 600 samples were collected across fourpopulations in Cameroon and five populations in DRC. There wasunclear geographic separation between collections in Cameroon andDRC, and these populations were therefore combined. It is likely thishappened due to incorrect recording of sample coordinates.In addition, a population from Kenya was included in the geneticanalysis to provide further comparison between populations.For more detail on the sample quantities and locations, refer toAppendix B. DNA sample records were provided in the mid-termproject report delivered December 2014.7

Activity 1.1.3 Genetic marker developmentProjectdocumentdescriptionThe samples from the above populations will be processed in thelaboratory to identify ‘polymorphic’ genetic markers. These are markersthat can be used to differentiate between individual trees of the samespecies.Changes toactivityWhilst the number of markers to be developed was not specified in theoriginal project description, it was decided to increase the number ofmarkers developed from 10 to 15 in order to build capability forpopulation assignment.Realized activityComplete. A total of 36 different Prunus SSR markers (31 nuclear and 5chloroplast SSR markers) were tested for amplification and segregationin the available Prunus africana populations. 16 markers (15 nuclearand 1 chloroplast SSR markers) were used in the final analysis.Based on these markers, it was possible to identify 9 distinct Prunusafricana populations from the samples submitted. All populations fromCameroon form one genetic cluster and all populations from East-Africa(DRC Kenya) form another cluster. Of interest is that the fourpopulations from Cameroon are split into two distinct genetic groups.The two populations (Cam Pop1 and Cam Pop4) in the South-West ofCameroon are very different from all the other populations.For more detail on the marker analysis, refer to Appendix B.Output 1.2: Capacity building and training of local teams in DNA sample collection andstorage.Activity 1.2.1 Workshops for stakeholder consultation and field trainingProjectdocumentdescriptionWe recommend an opening and closing workshop to inform, receiveinput and report back to key stakeholders involved in the ITTO-CITESproject.Suggested participant organisations include trade companyrepresentatives in Cameroon and DRC, government representatives,customs representatives and international buyers of Prunus bark.Changes toactivityRealized activityThe opening workshop will include a field trip to conduct training of staffin bark sampling techniques and can also be combined with thepopulation sampling (Activity 1.1.2).To save costs, we recommended combining workshops with the similarPericopsis elata project.Two opening workshops completed. Closing workshop completed inYaounde, Cameroon on 2 Feb 2016, with a further closing workshop inDRC planned for March 2016.In Cameroon we were able to combine the workshop and training forboth P. elata and Prunus africana projects. Agendas, attendance listsand photographs from the workshops have been provided in the midterm project report delivered December 2014.8

Output 1.3: Implementation of DNA verification in two controlled PAUs in Cameroonand DRC from pre-harvest to the factory and then to the exit points.Activity 1.3.1 Verification system developmentProjectdocumentdescriptionThis consists of identifying appropriate stages in the forestmanagement, harvesting and transportation process to take samplesfrom standing trees and bark.Care will be taken to work with concession and the national ITTOCITES project coordination teams to understand the field processes anddesign a sampling system that uses existing resources with minimumimpact on forest, transport and processing operations.Changes toactivityRealized activityAt the end of this stage a report will be presented explaining the systemdesign, statistical sampling strategy and resulting estimated level ofconfidence. (For example, “through this system we will be able to detectbark substitution amounting to 1% of total volume, 99% of the time”).No changeComplete. A number of phone calls and meetings have been held withmembers of ITTO, ANAFOR, MINFOF, MECNT and industry andacademic stakeholders. See Appendix C for a list of stakeholdermeetings held during the project.Final recommendations for a DNA-based verification system arepresented in Appendix D and are to be discussed in detail at theclosing meetings.Activity 1.3.2 Field implementationProjectdocumentdescriptionThe following basic process will be adapted according to the researchand findings made during Activity 1.3.1 Verification systemdevelopment.i)ii)iii)iv)Wood samples are taken from the cambium (the layer underneaththe bark of the tree) of Prunus africana trees either prior to harvestor during harvest.A second set of samples are taken from bark after harvest at anappropriate control point. Control points under consideration couldbe one or more of i) PAU exit gate ii) community drying or storageareas, iii) on delivery to processing facilities, iv) prior to loadingonto trucks for transport to the factories, v) at the port prior toexport or vi) on arrival at destination processing facilities inEurope.Using CoC documentation, the samples taken pre- and postharvest are physically matched together.A statistically representative number of paired samples areselected and sent to the DoubleHelix laboratory for DNA testing.The number of paired samples to be tested and the frequency of testingwill depend on the harvest plan and volume of timber harvested in eachperiod as well as the preferred timescales to identify problems andimplement corrective actions. DoubleHelix has previously applied ISO2859: Sampling procedures for inspection by attributes to calculate thenumber of samples to be tested according to the level of confidencerequired. A similar approach will be considered for this project.9

Changes toactivityThis has changed significantly due to the change in work scope towardpopulation analysis. Paired sampling and testing was conducted on aportion of samples in Cameroon and DRC. Another set of samples fromCameroon, DRC and Kenya were tested and assigned to population.Realized activitySamples for individual assignment were taken from one site inCameroon (AFRIMED), consisting of 55 pairs of cambium and bark, anda site in DRC (KAHINDO) consisting of 150 pairs of cambium and bark.Of these, 30 pairs from Cameroon and 60 pairs from DRC were tested.All remaining samples were collected from PAUs and harvest zones inDRC to be tested and assigned to population. For further detail, seeAppendix B.Activity 1.3.3 Ongoing DNA testingProjectdocumentdescriptionThe remainder of the DNA verification process is conducted in thelaboratory and does not impact forest operations.After DNA is extracted from the wood and bark samples, it is analysedand the genetic markers compared. If the DNA test confirms a geneticmatch between the paired samples, the chain-of-custody is provenintact. A genetic mis-match indicates that the paired samples come fromdifferent trees. This is evidence of substitution and further action can betaken.Changes toactivityThis activity was expanded to include both testing for individualassignment and population assignment.Realized activityBoth individual and population assignment tests were completed.Individual assignment revealed difficulties in successfully matchingmaterials from bark back to tree. Population assignment wascomparatively easy and more successful (see Activity 1.3.4). For furtherdetail see Appendix B.Activity 1.3.4 Statistical analysis of resultsProjectdocumentdescriptionA full analysis of the results of all the paired samples will be conductedand a final conclusion drawn on the likelihood of systematic substitutionwithin the supply chain. The test results and statistical analysis will beprovided in a report delivered regularly according to the frequency oftesting selected.Changes toactivityThis activity was expanded to include analysis of both individualassignment and population assignment.Realized activityIndividual assignmentThe practical test with material processed at a mill in Cameroon showeda high level of mis-match between results and chain of custody recordsof bark from individual trees. Thus we recommend focusing furtherapplications on verification of the population of harvest.Population assignmentGenetic differences between populations are highly significant and thesuccess of assignment back to the correct population was achieved witha mean average of 95%.For further detail see Appendix B.10

3.2 Comparison of activity progress against scheduleTable 2 below shows the project implementation in comparison with the detailed workplanprovided to ITTO in May 2014 and revised in March 2015. The approved workplan scheduleis marked in grey, with the actual implementation indicated by the hatched red cells.Table 2: Activity schedule in comparison with workplanOUTPUTS /ACTIVITIESRESPONSIBLE MayPARTY2014 JunMonthJanJul Aug Sep Oct Nov Dec 2015 Feb Mar Apr May JunJanJul Aug Sep Oct Nov Dec 2016 FebOutput 1.11.1.1 Trial DNAextractionDoubleHelix1.1.2 PopulationsamplingMECNT,ANAFOR,DoubleHelix1.1.3 DNA markerDoubleHelixdevelopmentOutput 1.21.2.1 Workshopsand trainingMECNT,ANAFOR,DoubleHelixOutput 1.31.3.1 VerificationsystemdevelopmentDoubleHelix1.3.2 FieldimplementationMECNT,ANAFOR,DoubleHelix1.3.3 OngoingDNA verificationDoubleHelix1.3.4 StatisticalanalysisDoubleHelix3.3 Project expenditure against funds input Total project expenditure to date by DoubleHelix: USD 223,716This figure consists:ooUSD 177,930 of direct expensesUSD 45,786 of project coordination and reporting costsThis figure does not include estimated expenses for attendance at the proposed projectclosing meeting in DRC in March 2016.The increased expenditure reflects the expansion in the scope of work and extension ofproject duration from 18 to 22 months. The additional cost has been absorbed by DoubleHelixat no charge to ITTO. For further detail on expenditure against budget (original and revised),see Appendix E. Total funds received from ITTO to date: USD 160,000 Remaining funds to be released: USD 36,10011

4PROJECT OUTCOMEThe objective of building capability to trace Prunus africana bark back to specific PAUs orharvest zones to a high level of confidence has been realised. The approach implemented inthis pilot project can be expanded to other authorized PAUs and harvest zones as part ofefforts to control and support the sustainable harvest and trade in Prunus bark.Attempts to sample and match Prunus bark from warehousing and processing facilities backto specific trees were challenging, even in cases where it was possible to segregate bark fromdifferent trees during harvest and transport.It is fortunate therefore, that the scope of work was adjusted to allow an implementation ofpopulation assignment to take place. Table 3 below clearly illustrates the effectiveness of thisapproach, where on average 95% of samples were assigned to the correct population.Table 3: Results of the self-assignment of individuals. The yellow cells mark thepercentage of correctly assigned individualsThere was some significant challenges to be overcome during the course of the project, whichis described in detail in Section 5.4.1 Tangible outputsA suite of genetic markers36 different Prunus SSR markers (31 nuclear and 5 chloroplast SSR markers) have beendeveloped, with 16 of them (15 nuclear and 1 chloroplast) used for the final project analysis.These markers can be used for both individual assignment (matching of bark back toindividual trees, as well as population assignment (matching of bark back to a geographicarea). As well as their utility for bark traceability, these markers have many other potentialresearch applications as described below.Genetic reference data for 8 distinct Prunus populations across Cameroon and DRCFor the first time, genetic profiles have been created for specific regions of Prunus africana.The data generated has an extremely long lifetime and will be valid for traceability purposesfor generations to come. The raw sequence data generated during this project can be sharedwith other research institutes and scientists for other analysis such as landscapereconstruction and climate change adaptation.12

Recommendations for a system to support verification of CITES applications, export andimport permits.As described in Appendix D, DoubleHelix has proposed a DNA verification system that canbe used to verify the accuracy of CITES export permit applications and the export/importpermits themselves at key control points in the supply chain.As forest management practices develop in PAUs and harvest zones under the ITTO-CITESprogramme, the DNA verification system can be adapted to monitor improvements in theevolving document chain of custody requirements.4.2 Participation of target beneficiaries Training of sampling teams in Cameroon and DRC on sampling techniques, samplepacking, storage and logistics. Participants in the training included those from localacademic institutions, universities, Government departments and Prunus exploitationindustry. Participation of industry in sampling process at PAUs and warehouse/distribution centres. Ongoing communication with Prunus exporters in DRC and importers in the EU. Discussion with CITES authorities on potential implementation of systems in Cameroon.Refer to Appendix C for details on meetings with stakeholders describing participationin the project.4.3 Next stepsAt the request of the European Commission SRG, a paper has been prepared by DoubleHelixon recommended activities to follow on from this project.In order for a traceability system to be of full benefit, it must be applied to the total coveragearea of Prunus exploitation. Our recommendations include: Peer review and publication of the scientific genetic marker development work. Additional sampling and marker development to enlarge the genetic reference databaseto include an additional 15 PAUs and harvest zones across the region. Formalisation of the procedures for sampling, testing and reporting in consultation withGovernment, the CITES Secretariat, industry and other stakeholders. The roll out of a screening programme (routine verification of CITES permits) to controlrisk of substitution from non-authorised areas, as described in Appendix D. Create a fast-track CITES permit issuing system for scientific samples to ensure speed ofdelivery of test samples from Cameroon and DRC to laboratories. Develop capacity to conduct testing within the region. As part of the ITTO Congo Basinproject, basic lab facilities have been set up with varying success in Kenya, Ghana andGabon. A proposal was submitted to ITTO in September 2014 for the creation of alaboratory in Cameroon.Note that the challenges of setting up a lab should not be underestimated. Once facilitiesand equipment have been installed, significant

Pilot Implementation of a DNA traceability system for Prunus africana in Prunus Allocation Units in Cameroon and Democratic Republic of Congo (DRC). . Project coordinator Double Helix Tracking Technologies Pte Ltd 105 Cecil Street, #22-00 The Octagon . Due to difficulties in bark segregation in DRC and ongoing security concerns, DoubleHelix .